Strong Relations Between ENSO and the Arctic Oscillation in the North American Multimodel Ensemble

Arctic Oscillation (AO) variability impacts climate anomalies over the middle to high latitudes of the Northern Hemisphere. Recently, state-of-the-art climate prediction models have proved capable of skillfully predicting the AO during the winter, revealing a previously unrealized source of climate predictability. Hindcasts from the North American Multimodel Ensemble (NMME) show that the seasonal, ensemble mean 200 hPa AO index is skillfully predicted up to 7 months in advance and that this skill, especially at longer leads, is coincident with previously unknown and strong relations (r > 0.9) with the El Nino–Southern Oscillation (ENSO). The NMME is a seasonal prediction system that comprises eight models and up to 100 members with forecasts out to 12 months. Observed ENSO-AO correlations are within the spread of the NMME member correlations, but the majority of member correlations are stronger than observed, consistent with too high predictability in the model, or overconfidence.